Intelligent regulation and control method and system for photovoltaic energy storage air conditioner

Abstract

The invention discloses an intelligent regulation and control method and system for a photovoltaic energy storage air conditioner. The method comprises the steps that photovoltaic related data, energy storage related data and air conditioner operation related data are collected; inputting the photovoltaic related data into the photovoltaic power generation prediction model to obtain a photovoltaic power generation prediction result; inputting the air conditioner operation related data into the air conditioner load prediction model to obtain an air conditioner load prediction result; inputting the photovoltaic generating capacity prediction result, the air conditioning load prediction result, the energy storage related data and the photovoltaic related data into an energy storage optimization scheduling model to obtain an energy storage charging and discharging decision; constructing a power supply strategy core rule base based on the energy storage related data, the electricity price peak time period and output results of the photovoltaic power generation prediction model and the air conditioner load prediction model; and optimizing an output result of the energy storage optimization scheduling model based on the power supply strategy core rule base. The method has the outstanding advantages of high reliability, high photovoltaic utilization rate, low dependence degree on a power grid and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic energy storage, and in particular to an intelligent control method and system for photovoltaic energy storage air conditioners. Background Art

[0002] As an important component of building energy management, photovoltaic energy storage air conditioning systems are rapidly developing towards intelligent and efficient systems. However, existing systems still have many deficiencies in operation and management, including:

[0003] There is a lack of a unified intelligent control system, and there is no effective coordinated optimization mechanism between photovoltaic power generation, energy storage systems, and air conditioning loads;

[0004] Energy scheduling strategies are fixed and lack dynamic response capabilities, making it difficult to adapt to real-time changes in operating conditions, resulting in low energy efficiency and energy waste.

[0005] Therefore, it is urgent to build a photovoltaic energy storage air conditioning control system that integrates multi-source data and has dynamic optimization capabilities to improve the overall performance of photovoltaic energy storage air conditioning.

[0006] CN119123578A discloses a heating, ventilation, and air conditioning (HVAC) automatic control system with intelligent control. First, a data acquisition module collects current environmental parameters, solar power generation, and user demand at intervals of T. Second, the collected data is fed into a trained convolutional neural network, which outputs future solar power generation, environmental parameter trends, user demand, and control signals. Third, a central controller sends a control signal to the HVAC system, adjusting the operating parameters of the air conditioning equipment based on the output power for the future period, thereby automatically regulating indoor temperature and humidity. Finally, according to the central controller's instructions, the system selects whether to use solar power or grid power to drive the air conditioning equipment during the future period. However, this patent relies solely on a single convolutional neural network prediction model and lacks mechanisms for multi-model fusion or modular modeling, resulting in limited prediction accuracy and generalization. Furthermore, the patent fails to incorporate multiple sources of data, such as energy storage system status and electricity price information, resulting in a control strategy with insufficient awareness of the system's overall operating status.

[0007] CN117091242A discloses an evaluation method, temperature setting method and control system for an air-conditioning temperature-controlled load cluster, the method comprising: obtaining environmental characteristics and cluster characteristics of the air-conditioning temperature-controlled load cluster, the environmental characteristics comprising ambient temperature and ambient humidity, the cluster characteristics comprising total cluster power, average indoor temperature and cluster set temperature; determining the adjustable potential of the air-conditioning temperature-controlled load cluster based on a combination of the environmental characteristics and cluster characteristics; wherein the adjustable potential of the air-conditioning temperature-controlled load cluster comprises: adjustable power amplitude and duration. However, this patent lacks comprehensive consideration of photovoltaic power generation and energy storage systems, and cannot achieve coordinated regulation of multiple energy sources. The power adjustment amplitude and duration are calculated based on environmental and cluster characteristics, lacking dynamic prediction and real-time optimization scheduling, resulting in delayed regulation decisions and making it difficult to adapt to complex and changing operating environments.

[0008] CN118935656A discloses a method, device, equipment, and medium for operating a photovoltaic air conditioner. The method includes: obtaining the current photovoltaic panel voltage, photovoltaic cell voltage, photovoltaic cell power, and battery voltage drop rate; determining a corresponding air conditioning parameter adjustment strategy based on the photovoltaic panel voltage, photovoltaic cell voltage, photovoltaic cell power, battery voltage drop rate, and a preset photovoltaic panel voltage, preset photovoltaic cell voltage, and preset photovoltaic cell power; and adjusting the operating parameters of the photovoltaic air conditioner according to the air conditioning parameter adjustment strategy to control air conditioning operation. However, the patent focuses on simple parameter adjustments of photovoltaic panel voltage, battery power, and voltage drop rate; the adjustment strategy is simple, adjusting the air conditioning parameters only according to the current state, without considering factors such as air conditioning load forecasting, energy storage scheduling, and electricity price fluctuations. The control strategy is relatively rough, resulting in poor adjustment results. Summary of the Invention

[0009] In order to solve the problems existing in the prior art such as poor prediction accuracy, poor perception of the overall operating status of the system, and insufficient intelligence of the photovoltaic energy storage air-conditioning system, the present invention provides a photovoltaic energy storage air-conditioning intelligent control method and system.

[0010] The present invention adopts the following technical solutions.

[0011] The present invention discloses a photovoltaic energy storage air conditioner intelligent control method, comprising:

[0012] Collecting photovoltaic energy storage air conditioning related data; the photovoltaic energy storage air conditioning related data includes photovoltaic related data, energy storage related data, and air conditioning operation related data;

[0013] Input photovoltaic related data into the photovoltaic power generation prediction model to obtain the photovoltaic power generation prediction result; input air conditioning operation related data into the air conditioning load prediction model to obtain the air conditioning load prediction result;

[0014] The photovoltaic power generation forecast results, air conditioning load forecast results, energy storage related data, and the historical photovoltaic power generation data and air conditioning load data within the set first time period are input into the energy storage optimization scheduling model to obtain energy storage charging and discharging decisions;

[0015] Based on energy storage-related data, peak electricity price periods, and the output results of the photovoltaic power generation forecast model and the air conditioning load forecast model, a power supply strategy core rule base is constructed; based on the power supply strategy core rule base, the output results of the energy storage optimization scheduling model are optimized.

[0016] More preferably,

[0017] The method further includes: constructing a compressor state adjustment fuzzy rule base based on set user preferences and temperature and humidity values ​​in air conditioning operation-related data; and implementing adaptive adjustment of the air conditioning compressor operation state through a fuzzy controller based on the compressor state adjustment fuzzy rule base.

[0018] More preferably,

[0019] Photovoltaic related data include photovoltaic DC side output voltage, photovoltaic DC side output current, photovoltaic module temperature, outdoor irradiance, outdoor temperature, outdoor humidity and wind speed;

[0020] Air conditioning operation related data includes compressor frequency, condenser temperature, evaporator temperature, indoor temperature, indoor humidity and power consumption;

[0021] Energy storage related data includes the voltage of each single cell, the temperature of each single cell, the charge and discharge current of the energy storage, the energy storage status, the charge status, the health status and the charge and discharge rate; the energy storage status includes energy storage charging, discharging and neither charging nor discharging.

[0022] More preferably,

[0023] The power supply strategy core rule base includes the first core rule, the second core rule and the third core rule;

[0024] Among them, the first core rule has a higher priority than the second core rule; the second core rule has a higher priority than the third core rule;

[0025] The first core rule is that when the predicted photovoltaic power generation is greater than or equal to the predicted air conditioning load, photovoltaic power is used to directly power the air conditioning, and the remaining power is used to charge the energy storage.

[0026] The second core rule is that when the sum of the predicted photovoltaic power generation and the energy stored in the energy storage is greater than or equal to the predicted air conditioning load, photovoltaic power is used directly to power the air conditioning, and energy storage discharge is used to supplement the power supply difference;

[0027] The third core rule is that when electricity prices are at peak times and the energy storage state of charge is greater than the set charge threshold, energy storage is forced to power the air conditioner.

[0028] More preferably,

[0029] When optimizing the output results of the energy storage optimization scheduling model, the preset energy storage protection strategy is executed;

[0030] The energy storage protection strategy includes energy storage system start-up and stall logic and minimum operating current protection.

[0031] More preferably,

[0032] The energy storage system start-up and shutdown logic includes:

[0033] If the energy storage needs to switch from charging to discharging and the first start-stop switching condition is met, the energy storage is discharged;

[0034] If the energy storage needs to switch from discharging to charging and the second start-stop switching condition is met, the energy storage is charged.

[0035] More preferably,

[0036] The first switching condition after the start-up stagnation is to satisfy the following formula within the set anti-shake delay:

[0037] P c <P c_on_threshold +P hys

[0038] Among them, P c is the actual charging power of energy storage; P c_on_threshold The required energy storage charging power; P hys is the hysteresis interval.

[0039] More preferably,

[0040] The second switching condition after the start-up and stagnation is to satisfy the following formula within the set anti-shake delay:

[0041] P l ≥P l_on_threshold +P hys

[0042] Among them, P l is the actual discharge power of energy storage; P l_on_threshold is the required energy storage discharge power.

[0043] More preferably,

[0044] The minimum operating current protection is to perform charging and discharging only when the energy storage charging and discharging current is greater than or equal to the set minimum operating current threshold, otherwise it remains idle.

[0045] More preferably,

[0046] The compressor state adjustment fuzzy rule base includes:

[0047] When the temperature error is greater than a set first temperature error threshold, the error change rate is greater than a set first change rate threshold, and the indoor humidity is greater than the user-set humidity, the air conditioner compressor speed is reduced to a set first speed; the temperature error is the difference between the user-set temperature and the indoor temperature; the error change rate is the rate of change of the temperature error;

[0048] When the temperature error is negative and greater than the set second temperature error threshold, and the error change rate is less than the set second change rate threshold, and the indoor humidity is less than the user-set humidity, the air conditioner compressor speed is increased to the set second speed;

[0049] When the temperature error is negative and less than a set second temperature error threshold, increasing the speed of the air-conditioning compressor to a set third speed;

[0050] In other cases, the current air conditioner compressor speed is maintained;

[0051] a first temperature error threshold; the first temperature error threshold and the first change rate threshold are both positive numbers;

[0052] The second temperature error threshold and the second change rate threshold are both negative numbers; the ratio of the second speed to the current air-conditioning compressor speed is less than the ratio of the third speed to the current air-conditioning compressor speed.

[0053] On the other hand, the present invention discloses a photovoltaic energy storage air conditioning intelligent control system based on the photovoltaic energy storage air conditioning intelligent control method, including a multi-source data acquisition module, a prediction and optimization modeling module and an intelligent control module:

[0054] The multi-source data acquisition module collects photovoltaic energy storage air conditioning related data; the photovoltaic energy storage air conditioning related data includes photovoltaic related data, energy storage related data, and air conditioning operation related data;

[0055] The prediction and optimization modeling module inputs photovoltaic-related data into the photovoltaic power generation prediction model to obtain photovoltaic power generation prediction results; inputs air conditioning operation-related data into the air conditioning load prediction model to obtain air conditioning load prediction results; and inputs photovoltaic power generation prediction results, air conditioning load prediction results, energy storage-related data, and photovoltaic power generation historical data and air conditioning load historical data within a set first time period into the energy storage optimization scheduling model to obtain energy storage charging and discharging decisions;

[0056] The intelligent control module constructs a power supply strategy core rule base based on energy storage-related data, peak electricity price periods, and the output results of the photovoltaic power generation prediction model and the air conditioning load prediction model; optimizes the output results of the energy storage optimization scheduling model based on the power supply strategy core rule base; and constructs a compressor state adjustment fuzzy rule base based on the set user preferences and temperature and humidity values ​​in the air conditioning operation-related data; and uses the compressor state adjustment fuzzy rule base to achieve adaptive adjustment of the air conditioning compressor operating state through a fuzzy controller.

[0057] The present application also discloses a computer-readable storage medium having a computer program stored thereon, wherein the computer program is characterized in that when the program is executed by a processor, the photovoltaic energy storage air conditioner intelligent control method is implemented.

[0058] The beneficial effects of the present invention are as follows:

[0059] The present invention integrates multi-dimensional data sources such as photovoltaic power generation, energy storage status, air conditioning operation, indoor and outdoor environment, and user behavior, and realizes local preprocessing and exception elimination of data through edge computing nodes, thereby improving data quality and response efficiency, and providing highly reliable data support for subsequent modeling and control.

[0060] The present invention also constructs a power generation and load forecasting model with XGBoost and temporal convolutional network as the core, combines dynamic electricity prices, battery health status and other information, and realizes dynamic optimization scheduling of the energy storage system based on reinforcement learning algorithm, supporting feedforward energy optimization management of the photovoltaic storage air-conditioning system, improving photovoltaic utilization and overall system energy efficiency, and reducing dependence on the power grid.

[0061] Based on the control logic that combines state perception and rule-driven control, the present invention designs an energy flow adaptive decision-making engine to implement a layered energy supply strategy with photovoltaic priority direct supply, energy storage auxiliary regulation, and intelligent grid backup. It also integrates fuzzy control algorithms to achieve intelligent regulation of air-conditioning compressor operation, implement user behavior-oriented intelligent control strategies, and dynamically balance user comfort and energy efficiency goals.

[0062] In summary, this invention addresses the insufficient intelligence of existing photovoltaic energy storage and air conditioning systems by proposing an intelligent control system that integrates multi-source data acquisition, predictive modeling, and intelligent regulation. Based on an edge-cloud collaborative architecture and incorporating predictive control algorithms, this system achieves coordinated and optimized control of photovoltaic power generation, energy storage systems, and air conditioning operations, effectively improving overall energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of the intelligent control method for photovoltaic energy storage air conditioners;

[0064] Figure 2This is a schematic diagram of the photovoltaic energy storage air conditioning intelligent control system module. DETAILED DESCRIPTION

[0065] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without making creative efforts are all within the scope of protection of the present invention.

[0066] This application discloses a photovoltaic energy storage air conditioner intelligent control method, see the attached Figure 1 ,include:

[0067] Collecting photovoltaic energy storage air conditioning related data; the photovoltaic energy storage air conditioning related data includes photovoltaic related data, energy storage related data, and air conditioning operation related data;

[0068] Photovoltaic related data include photovoltaic DC side output voltage, photovoltaic DC side output current, photovoltaic module temperature, outdoor irradiance, outdoor temperature, outdoor humidity and wind speed;

[0069] Air conditioning operation related data includes compressor frequency, condenser temperature, evaporator temperature, indoor temperature, indoor humidity and power consumption;

[0070] Energy storage related data includes the voltage of each single cell, the temperature of each single cell, the charge and discharge current of the energy storage, the energy storage status, the charge status, the health status and the charge and discharge rate; the energy storage status includes energy storage charging, discharging and neither charging nor discharging.

[0071] Input photovoltaic related data into the photovoltaic power generation prediction model to obtain the photovoltaic power generation prediction result; input air conditioning operation related data into the air conditioning load prediction model to obtain the air conditioning load prediction result;

[0072] Inputting photovoltaic power generation forecast results, air conditioning load forecast results, energy storage related data, and historical photovoltaic power generation data and air conditioning load data within a set first time period into the energy storage optimization scheduling model to obtain energy storage charging and discharging decisions; preferably, the first time period is set to 1 day;

[0073] Based on energy storage-related data, peak electricity price periods, and the output results of the photovoltaic power generation forecast model and the air conditioning load forecast model, a power supply strategy core rule base is constructed; based on the power supply strategy core rule base, the output results of the energy storage optimization scheduling model are optimized.

[0074] The power supply strategy core rule base includes the first core rule, the second core rule and the third core rule;

[0075] Among them, the first core rule has a higher priority than the second core rule; the second core rule has a higher priority than the third core rule;

[0076] The first core rule is that when the predicted photovoltaic power generation is greater than or equal to the predicted air conditioning load, photovoltaic power is used to directly power the air conditioning, and the surplus power after powering the air conditioning is used to charge the energy storage;

[0077] The second core rule is that when the sum of the predicted photovoltaic power generation and the power stored in the energy storage exceeds the predicted air conditioning load, photovoltaic power will be used directly to power the air conditioning, and energy storage will be used to power the air conditioning to make up for the difference between the photovoltaic power generation and the air conditioning load.

[0078] The third core rule is that when electricity prices are at peak times and the energy storage state of charge is greater than the set charge threshold, energy storage is forced to power the air conditioner.

[0079] When optimizing the output results of the energy storage optimization scheduling model, the preset energy storage protection strategy is executed;

[0080] The energy storage protection strategy includes energy storage system start-up and stall logic and minimum operating current protection.

[0081] The energy storage system start-up and shutdown logic includes:

[0082] If the energy storage needs to switch from charging to discharging and the first start-stop switching condition is met, the energy storage is discharged;

[0083] The first switching condition after the start-up stagnation is to satisfy the following formula within the set anti-shake delay:

[0084] P c <P c_on_threshold +P hys

[0085] Among them, P c is the actual charging power of energy storage; P c_on_threshold The required energy storage charging power; P hys Preferably, the required energy storage charging power is -0.5kW; the preferred value range of the hysteresis interval is ±0.2kW, charging is negative and discharging is positive; the anti-shake delay is set to be less than or equal to 30 seconds.

[0086] If the energy storage needs to switch from discharging to charging and the second start-stop switching condition is met, the energy storage is charged.

[0087] The second switching condition after the start-up and stagnation is to satisfy the following formula within the set anti-shake delay:

[0088] P l ≥P l_on_threshold +P hys

[0089] Among them, P l is the actual discharge power of energy storage; P l_on_threshold is the required energy storage discharge power. Preferably, the required energy storage discharge power is 0.5kW;

[0090] The minimum operating current protection is to perform charging and discharging only when the energy storage charging and discharging current is greater than or equal to the set minimum operating current threshold, otherwise it remains idle.

[0091] The method further includes: constructing a compressor state adjustment fuzzy rule base based on user preferences and temperature and humidity values ​​in air conditioner operation-related data; and adaptively adjusting the operating state of the air conditioner compressor using a fuzzy controller based on the compressor state adjustment fuzzy rule base. The user preferences include the temperature and humidity set by the user.

[0092] The compressor state adjustment fuzzy rule base includes:

[0093] When the temperature error is greater than a set first temperature error threshold, the error change rate is greater than a set first change rate threshold, and the indoor humidity is greater than the user-set humidity, the air conditioner compressor speed is reduced to a set first speed; the temperature error is the difference between the user-set temperature and the indoor temperature; the error change rate is the rate of change of the temperature error;

[0094] When the temperature error is negative and greater than the set second temperature error threshold, and the error change rate is less than the set second change rate threshold, and the indoor humidity is less than the user-set humidity, the air conditioner compressor speed is increased to the set second speed;

[0095] When the temperature error is negative and less than a set second temperature error threshold, increasing the speed of the air-conditioning compressor to a set third speed;

[0096] In other cases, the current air conditioner compressor speed is maintained;

[0097] The first temperature error threshold; the first temperature error threshold and the first rate of change threshold are both positive numbers; those skilled in the art can set the first temperature error threshold and the first rate of change threshold based on actual conditions, and will not be further described here. The set first speed is based on the current air conditioner compressor speed. Those skilled in the art can set the first speed to be less than the current air conditioner compressor speed based on actual conditions, and will not be further described here.

[0098] The second temperature error threshold and the second rate of change threshold are both negative numbers; the ratio of the second speed to the current air conditioner compressor speed is less than the ratio of the third speed to the current air conditioner compressor speed. Those skilled in the art can set the second temperature error threshold and the second rate of change threshold based on actual circumstances, and this is not further described here. The second and third speeds are set based on the current air conditioner compressor speed, and those skilled in the art can set the second and third speeds based on actual circumstances, and this is not further described here.

[0099] Example 1

[0100] A photovoltaic energy storage air conditioner intelligent control method, see the attached Figure 1 ,include:

[0101] Step 1: Collect relevant data of photovoltaic energy storage air conditioner and pre-process the collected data;

[0102] The collected photovoltaic energy storage air conditioning related data includes photovoltaic related data, energy storage related data, and air conditioning operation related data;

[0103] The photovoltaic related data is obtained by deploying a high-precision sensor group at the key nodes of the photovoltaic array to collect the photovoltaic DC side output voltage (U pv ), photovoltaic DC side output current (I pv ), PV module temperature (T pv ), outdoor irradiance, air temperature, humidity, and wind speed, among other core parameters. The components include photovoltaic panels and inverters; the key nodes may be the output of the photovoltaic panels and the input and output of the inverters. Preferably, the sampling period is set to 5 seconds.

[0104] The energy storage-related data is collected in a refined manner using a three-level monitoring architecture. At the battery pack level, voltage and temperature sensors are installed at each cell to collect the voltage, temperature, and other status of each cell. At the module level, Hall effect current sensors are configured to collect the charge and discharge currents of the energy storage. At the system level, a battery management system (BMS) controller is deployed to obtain key operating indicators such as the state of charge (SOC), state of health (SOH), charge and discharge rate (C-rate), and energy storage status. Energy storage status includes charging, discharging, and neither charging nor discharging.

[0105] The air-conditioning operation related data includes the compressor frequency (F comp ), condenser temperature (T cond ), evaporator temperature (T evap ), indoor air temperature, humidity, power consumption data and other operating parameters.

[0106] Preprocessing is performed on various types of collected data, including time synchronization, data format unification, and outlier removal.

[0107] The collected data is transmitted to the edge gateway via the LoRa long-distance radio wireless communication network, and then uploaded to the cloud platform stably and efficiently using the MQTT message queue telemetry transmission protocol, realizing remote aggregation and management of system-wide data.

[0108] Step 2: Perform data cleaning and build a multidimensional data prediction model based on the machine learning model;

[0109] Step 2.1: Data cleaning;

[0110] Data cleaning involves cleaning and improving the quality of raw data. First, the Kalman filter algorithm is used to remove sensor noise, improving data stability and accuracy. Linear interpolation is then used to properly fill in missing data caused by sensor failure or communication interruptions, restoring data integrity. Three-sigma anomaly detection is used to identify and remove outliers to ensure the quality of modeled data.

[0111] Step 2.2: Build a multidimensional data prediction model based on the machine learning model;

[0112] The multidimensional data prediction model includes a photovoltaic power generation prediction model, an air conditioning load prediction model, and an energy storage optimization scheduling model;

[0113] 2.2.1 Constructing a photovoltaic power generation prediction model based on the first machine learning model;

[0114] Preferably, the first machine learning model can adopt the extreme gradient boosting algorithm XGBoost, recurrent neural network RNN, etc.; those skilled in the art can choose according to actual conditions. In a preferred embodiment of the present invention, a photovoltaic power generation prediction model is constructed based on the XGBoost algorithm.

[0115] The photovoltaic power generation prediction model combines the real-time outdoor air temperature, outdoor air humidity, irradiance, outdoor wind speed and other meteorological parameters and the photovoltaic DC side output voltage (U pv ), photovoltaic DC side output current (I pv ), PV module temperature (T pv ) and other historical power generation data, and using the XGBoost algorithm to predict the photovoltaic power generation within a first time period set in the future; preferably, the first time period set is 24 hours;

[0116] Specifically, a photovoltaic power generation prediction model is constructed based on the XGBoost algorithm;

[0117] The outdoor air temperature, outdoor air humidity, irradiance, outdoor wind speed and other meteorological parameters and the photovoltaic DC side output voltage (U pv ), photovoltaic DC side output current (I pv ), PV module temperature (T pv ) as the input data of the photovoltaic power generation prediction model, and the photovoltaic power generation in the first period after the current moment as the output data of the photovoltaic power generation prediction model for training;

[0118] The real-time collected outdoor air temperature, outdoor air humidity, irradiance, outdoor wind speed and other meteorological parameters and the photovoltaic DC side output voltage (U pv ), photovoltaic DC side output current (I pv ), PV module temperature (T pv ) and other power generation data are input into the trained photovoltaic power generation prediction model to obtain the predicted photovoltaic power generation in the first period of the future, providing prior information for energy storage scheduling.

[0119] 2.2.2 Build an air conditioning load prediction model based on the second machine learning model;

[0120] Preferably, the second machine learning model can adopt a temporal convolutional network (TCN), a long short-term memory network (LSTM), etc. Those skilled in the art can select the model based on actual conditions. In a preferred embodiment of the present invention, an air conditioning load prediction model is constructed based on the TCN algorithm.

[0121] The air conditioning load prediction model uses the time convolution network (TCN) model based on indoor and outdoor temperature, humidity, compressor frequency, condenser temperature, evaporator temperature, operating mode, power consumption data, etc. to output the air conditioning load prediction results for the preset second period and the preset third period; preferably, the preset second period is 15 minutes; the preset third period is 4 hours

[0122] Specifically, an air conditioning load prediction model is constructed based on the TCN model. The input data for the air conditioning load prediction model include indoor and outdoor temperature, humidity, user-set temperature and humidity, PV DC side voltage and current, component temperature, compressor frequency, condenser temperature, evaporator temperature, operating mode, and power consumption data. The air conditioning loads for the second and third periods after the current moment are used as the output data for training.

[0123] The real-time collected indoor and outdoor temperature, humidity, user-set temperature and humidity, photovoltaic DC side voltage, current, component temperature, compressor frequency, condenser temperature, evaporator temperature, operating mode, and power consumption data are input into the trained air-conditioning load prediction model to obtain the predicted air-conditioning load forecast results for the second and third time periods in the future, assisting in formulating accurate power consumption control strategies.

[0124] 2.2.3 Construct a photovoltaic power generation prediction model based on the third machine learning model;

[0125] Preferably, the second machine learning model can adopt a reinforcement learning algorithm (such as an online strategy On-Policy algorithm, a deep Q network DQN, etc.); those skilled in the art can make a choice according to actual conditions.

[0126] The energy storage optimization scheduling model integrates photovoltaic power generation forecast results, air conditioning load forecast results, historical photovoltaic system power generation data, historical air conditioning system load data, the voltage and temperature of each single cell of the energy storage, the charge and discharge current of the energy storage, the real-time energy storage system operating status, state of charge (SOC), state of health (SOH), charge and discharge rate (C-rate) and peak and valley electricity price information. Based on reinforcement learning algorithm training, it outputs energy storage charging and discharging strategies that take into account both economy and stability, and realize dynamic optimization scheduling of the energy storage system.

[0127] Specifically, the energy storage optimization scheduling model is trained based on a reinforcement learning algorithm. It constructs a state space that includes historical data on photovoltaic power generation, historical data on air conditioning load, real-time energy storage system operating status (energy storage system operating status includes energy storage charging, energy storage discharging, and energy storage neither charging nor discharging), the voltage and temperature of each single cell, the energy storage charge and discharge current, state of charge (SOC), state of health (SOH), charge and discharge rate (C-rate), and peak and valley electricity prices, as well as an action space for energy storage charge and discharge power decisions (decision-making on the subsequent energy storage system operating status). The state space is used as the input of the energy storage optimization scheduling model, and the action space is used as the output of the action space. Through the objective function of maximizing long-term comprehensive benefits, the optimal energy storage scheduling strategy is learned and output, thereby taking into account both economic efficiency and system stability, and effectively improving the overall operating efficiency of the system.

[0128] The real-time collected photovoltaic power generation prediction results, air conditioning load prediction results, the voltage and temperature of each single cell of the energy storage, the charge and discharge current of the energy storage, the state of charge (SOC), the state of health (SOH), the charge and discharge rate (C-rate) and the peak and valley electricity price are input into the trained energy storage optimization scheduling model to obtain the energy storage scheduling strategy.

[0129] Step 3: Based on the output of each prediction model and user preferences, multi-source data and edge computing capabilities are integrated to achieve intelligent decision-making on energy flow paths, adaptive adjustment of air conditioning operating status, and continuous optimization of the overall system operating efficiency.

[0130] 3.1 Build a state-aware and rule-driven energy flow allocation rules engine based on multi-source data. The rules engine adjusts the power supply strategy according to a preset core rule base; the core rule base is built based on the collected data and the prediction results of the multi-dimensional data prediction optimization model;

[0131] By dynamically analyzing multidimensional time-series data such as photovoltaic power generation, air conditioning load, electricity price signals, and energy storage status, a state-aware and rule-driven energy flow allocation rules engine was constructed. This engine adjusts energy path priorities according to the predefined rules in the core rule library in Table 1, implementing a tiered energy supply strategy that prioritizes direct photovoltaic power generation, balances energy storage, and provides a backup grid.

[0132] Table 1 Core rule base

[0133]

[0134] The photovoltaic output refers to the photovoltaic DC side output power; preferably, the charging threshold is 50%;

[0135] The startup energy storage protection strategy is introduced to set the energy storage system start-stop logic and minimum working current protection to prevent frequent start-stop and system life degradation.

[0136] The energy storage system start-up and shutdown logic includes:

[0137] If the energy storage needs to switch from charging to discharging and the first start-stop switching condition is met, the energy storage is discharged; the first start-stop switching condition is to meet the following formula within the set anti-shake delay:

[0138] P c <P c_on_threshold +P hys

[0139] Among them, P c is the actual charging power of energy storage; P c_on_threshold The required energy storage charging power; P hys is the hysteresis interval; preferably, the required energy storage charging power is -0.5kW; the preferred value range of the hysteresis interval is ±0.2kW, charging is negative and discharging is positive; the anti-shake delay is set to less than or equal to 30 seconds.

[0140] If the energy storage needs to switch from discharging to charging and the second start-stop switching condition is met, the energy storage is charged.

[0141] The second switching condition after the start-up and stagnation is to satisfy the following formula within the set anti-shake delay:

[0142] P l ≥P l_on_threshold +P hys

[0143] Among them, P l is the actual discharge power of energy storage; P l_on_threshold is the required energy storage discharge power; P hysis the hysteresis interval; preferably, the required energy storage discharge power is 0.5kW; the preferred value range of the hysteresis interval is ±0.2kW, charging is negative and discharging is positive; the anti-shake delay is set to less than or equal to 30 seconds.

[0144] The first start-stop switching condition and the second start-stop switching condition are both set based on the actual battery power and the required battery power.

[0145] The minimum operating current protection is to only perform charging and discharging when the energy storage charging and discharging current is greater than or equal to the set minimum operating current threshold, otherwise it remains idle. The function of the minimum operating current protection is to eliminate small current operations and reduce battery loss. Therefore, those skilled in the art should be able to set the minimum operating current threshold according to the actual energy storage capacity; for example, when the energy storage capacity is 20Ah, the minimum operating current threshold is set to 2A; when the energy storage capacity is 30Ah, the minimum operating current threshold is set to 3A.

[0146] Based on the scheduling execution agent mechanism of the edge gateway, the power supply strategy is adjusted periodically according to the set first time interval to achieve real-time dynamic adjustment of energy flow. Preferably, the set first time interval is 1 minute;

[0147] Those skilled in the art should understand that the edge gateway-based scheduling and execution agent mechanism transmits prediction results to the edge gateway via the MQTT protocol. The edge gateway's scheduling and execution agent mechanism is a real-time decision-making and control module deployed on the edge gateway. It dynamically adjusts energy flow paths (such as the power supply ratio of photovoltaics, energy storage, and the power grid) and issues execution instructions. The edge gateway's scheduling and execution agent mechanism does not require training, and its dynamic adjustment capability is achieved through a rules engine and real-time data.

[0148] 3.2 Based on the fuzzy controller and the preset fuzzy rule base, adaptive adjustment of the operating state of the air-conditioning compressor is achieved; the preset fuzzy rule base is set based on user preferences and environment-related data.

[0149] By integrating multi-source environmental and operational data and combining them with predictive model results, intelligent and efficient air-conditioning control can be achieved. Specifically, the fuzzy PID controller uses "temperature error," "error change rate," and "humidity feedback" as fuzzy input variables, and utilizes fuzzy logic reasoning and a rule base to generate control outputs. The control signal is ultimately converted into a PWM duty cycle, which is used to adjust the compressor speed and achieve fine-tuning of the cooling or heating intensity. The fuzzy set partitioning and rule design in the controller are optimized based on the actual operating conditions of the system and user experience requirements to ensure a smooth and responsive adjustment process. The temperature error is the instantaneous deviation between the temperature value set by the user and the actual measured indoor temperature; the error change rate represents the rate of change of the temperature error over time; and the humidity feedback is the deviation between the humidity value set by the user and the actual measured humidity.

[0150] The preset fuzzy rule base includes:

[0151] When the temperature error is greater than a set first temperature error threshold, the error change rate is greater than a set first change rate threshold, and the indoor humidity is greater than the user-set humidity, the air conditioner compressor speed is reduced to a set first speed; the temperature error is the difference between the user-set temperature and the indoor temperature; the error change rate is the rate of change of the temperature error;

[0152] When the temperature error is less than the set second temperature error threshold, the error change rate is 0, and the indoor humidity is equal to the user-set humidity, the current air-conditioning compressor speed is maintained;

[0153] When the temperature error is negative and greater than the set third temperature error threshold, and the error change rate is greater than the set second change rate threshold, and the indoor humidity is lower than the user-set humidity, the air conditioner compressor speed is increased to the set second speed;

[0154] When the temperature error is negative and less than a set fourth temperature error threshold, and the error change rate is less than a set second change rate threshold, increasing the air conditioner compressor speed to a set third speed;

[0155] The first temperature error threshold is greater than the second temperature error threshold; the first temperature error threshold, the second temperature error threshold, and the first rate of change threshold are all positive numbers; those skilled in the art can set the first temperature error threshold, the second temperature error threshold, and the first rate of change threshold based on actual conditions, and will not be described in detail here. The set first speed is based on the current air conditioner compressor speed. Those skilled in the art can set the first speed to be less than the current air conditioner compressor speed based on actual conditions, and will not be described in detail here.

[0156] The third temperature error threshold is greater than the fourth temperature error threshold; the third temperature error threshold, the fourth temperature error threshold, and the second rate of change threshold are all negative; the ratio of the second speed to the current air-conditioning compressor speed is less than the ratio of the third speed to the current air-conditioning compressor speed. Those skilled in the art can set the third temperature error threshold, the fourth temperature error threshold, and the second rate of change threshold based on actual conditions, and will not be described in detail here. The set second and third speeds are based on the current air-conditioning compressor speed. Those skilled in the art can set the second and third speeds based on actual conditions, and will not be described in detail here.

[0157] 3.3 Switch the operating mode in real time based on the mode switching coefficient calculated in real time.

[0158] The operating modes include energy-saving mode and comfort mode; in energy-saving mode, the error range between indoor temperature control and the user-set preferred temperature is allowed to be ±3°C, reducing the number of frequent starts and stops of the compressor, thereby reducing power consumption and equipment wear, and extending system life. In comfort mode, the error range between temperature control and the user-set preferred temperature is narrowed to ±1°C to enhance the user's comfort experience, which is suitable for scenarios with high requirements for indoor environment. Mode switching can be automatically executed based on factors such as the current operating mode set by the user, the comfort mode time period set by the user, the power consumption budget and the air-conditioning load, so as to achieve a dynamic balance between power consumption and comfort. Specifically, first, weights are assigned to the current operating mode, power consumption budget, air-conditioning load and comfort mode time period set by the user, the mode switching coefficient is calculated, and the operating mode is determined according to the mode switching coefficient; the mode switching coefficient C O As shown in the following formula:

[0159]

[0160] Among them, a is the weight of the operating mode; R u The current operating mode set by the user. When the current operating mode is energy saving mode, R u When the current operating mode is comfort mode, R u is 2; b is the power consumption budget weight; E b The energy consumption budget for the second time period is set (in kW.h); E n is the power consumption of the air conditioner in the first set period (in kW.h); preferably, the second period is set to 1 day; c is the air conditioner load weight; A l is the air conditioning load (in kW); A e is the rated load of the air conditioner (in kW); d is the weight of the comfort mode time period; T s The length of comfort mode time set by the user, in h; T e The first time length standard is set, preferably, the first time length standard is 24 hours.

[0161] The operating mode weight, power consumption budget weight, air conditioning load weight, and comfort mode time period weight must meet the following constraints:

[0162]

[0163] Preferably, the weight of the operating mode is 0.4; the weight of the power consumption budget is 0.3; the weight of the air-conditioning load is 0.2; and the weight of the comfort mode time period is 0.1.

[0164] When the mode switching coefficient is greater than the set switching coefficient threshold, the comfort mode is adopted, otherwise, the energy-saving mode is adopted; those skilled in the art should be able to set the switching coefficient threshold according to actual conditions, which will not be repeated here.

[0165] 3.4 Further adjust the control strategy based on the air conditioning load forecast results.

[0166] For example, before a load peak is expected, the compressor's operating intensity is increased in advance to avoid frequent high-load starts and stops, improving system response flexibility and overall energy efficiency. When strong sunlight is predicted in the afternoon (based on the weather forecast), the system will combine photovoltaic output with the air conditioning load forecast to moderately pre-cool the indoor environment in the morning, allowing the air conditioning system to complete pre-cooling before the heat load surges, significantly reducing the pressure on electricity consumption under peak load. Those skilled in the art can determine the proportion of increased compressor operating intensity based on actual conditions, and can also determine the minimum temperature for pre-cooling the indoor environment based on the actual operating mode and the temperature and humidity set by the user. This will not be repeated here.

[0167] This application also discloses a photovoltaic energy storage air conditioning intelligent control system based on the photovoltaic energy storage air conditioning intelligent control method, see the attached Figure 2 , including multi-source data acquisition module, prediction and optimization modeling module and intelligent control module.

[0168] The multi-source data acquisition module collects photovoltaic energy storage air conditioning related data; the photovoltaic energy storage air conditioning related data includes photovoltaic related data, energy storage related data, and air conditioning operation related data;

[0169] The prediction and optimization modeling module inputs photovoltaic-related data into the photovoltaic power generation prediction model to obtain photovoltaic power generation prediction results; inputs air conditioning operation-related data into the air conditioning load prediction model to obtain air conditioning load prediction results; and inputs photovoltaic power generation prediction results, air conditioning load prediction results, energy storage-related data, and photovoltaic power generation historical data and air conditioning load historical data within a set first time period into the energy storage optimization scheduling model to obtain energy storage charging and discharging decisions;

[0170] The intelligent control module constructs a power supply strategy core rule base based on energy storage-related data, peak electricity price periods, and the output results of the photovoltaic power generation prediction model and the air conditioning load prediction model; optimizes the output results of the energy storage optimization scheduling model based on the power supply strategy core rule base; and constructs a compressor state adjustment fuzzy rule base based on the set user preferences and temperature and humidity values ​​in the air conditioning operation-related data; and uses the compressor state adjustment fuzzy rule base to achieve adaptive adjustment of the air conditioning compressor operating state through a fuzzy controller.

[0171] Example 2

[0172] A photovoltaic energy storage air conditioning intelligent control system, see attached Figure 2 , including multi-source data acquisition module, prediction and optimization modeling module and intelligent control module.

[0173] The multi-source data acquisition module includes a photovoltaic system acquisition unit, an energy storage system acquisition unit, an air-conditioning system acquisition unit, an environmental parameter acquisition unit, an edge computing unit, and a data transmission unit. It aims to comprehensively collect various types of raw data required for the operation of the photovoltaic energy storage and air-conditioning system, and provide high-quality data support for subsequent data modeling and intelligent control.

[0174] The photovoltaic system acquisition unit deploys a high-precision sensor group at the key nodes of the photovoltaic array to collect the photovoltaic DC side output voltage (U pv ), current (I pv ), component temperature (T pv ) and other core parameters. The components include photovoltaic panels, inverters, etc. The key nodes may be the output of the photovoltaic panels and the input and output of the inverters. Preferably, the set sampling period is 5 seconds.

[0175] The energy storage system data collection unit uses a three-level monitoring architecture to achieve refined data collection for the energy storage system. Voltage and temperature sensors are installed at the battery pack level to collect single cell voltage, temperature, and other status information. Hall effect current sensors are configured at the module level to collect charge and discharge currents. At the system level, a battery management system (BMS) controller is deployed to obtain key operating indicators such as state of charge (SOC), state of health (SOH), and charge and discharge rate (C-rate).

[0176] The air conditioning system acquisition unit collects the compressor frequency (F comp ), condenser temperature (T cond ), evaporator temperature (T evap ) and other operating parameters, as well as collecting the temperature and humidity set by the user, while linking the indoor environment sensors to obtain temperature and humidity data in real time to accurately feedback the user's load demand.

[0177] The environmental parameter collection unit deploys meteorological sensing devices outside the building to collect meteorological parameters such as irradiance, air temperature and humidity, and wind speed in real time, builds a local micro-meteorological database, and provides environmental input for photovoltaic power generation prediction and air conditioning load prediction.

[0178] Edge computing unit, the edge computing node performs local preprocessing on various types of collected data, including time synchronization, data format unification and outlier elimination.

[0179] The data transmission unit transmits the data processed by edge computing to the edge gateway through the wireless communication network of LoRa long-distance radio, and then uploads it to the cloud platform stably and efficiently using the MQTT message queue telemetry transmission protocol, realizing remote aggregation and management of data of the entire system.

[0180] The prediction and optimization modeling module includes a data cleaning unit, a photovoltaic power generation prediction model, an air conditioning load prediction model, and an energy storage scheduling prediction model; it is used to clean and process the collected data and perform modeling analysis to form the core basis for prediction and optimization.

[0181] The data cleaning unit cleans and improves the raw data. It first uses the Kalman filter algorithm to remove sensor noise, improving data stability and accuracy. It then uses linear interpolation to properly fill in missing data caused by sensor failure or communication interruptions to restore data integrity. It also uses the 3σ anomaly detection method to identify and remove outliers to ensure the quality of the modeled data.

[0182] Photovoltaic power generation prediction model, which combines real-time environmental parameters (irradiance, temperature, humidity, etc.) with historical power generation data, uses the XGBoost algorithm to predict photovoltaic power generation in the next 24 hours, providing prior information for energy storage scheduling.

[0183] The air conditioning load forecasting model, based on the building's internal and external environmental information and historical air conditioning operating condition data, uses a temporal convolutional network (TCN) to fully capture the time dependence of load demand, outputting short-term (15-minute level) and medium-term (4-hour level) air conditioning load forecast results to assist in formulating accurate electricity consumption control strategies.

[0184] The energy storage optimization scheduling model uses photovoltaic power generation forecasts, air conditioning load forecasts, historical photovoltaic system power generation data, historical air conditioning system load data, and the real-time energy storage system operating status as input. It comprehensively considers key operating parameters of the energy storage system, including state of charge (SOC), state of health (SOH), and charge and discharge rate (C-rate), and also incorporates peak and valley electricity price information. The model is trained using reinforcement learning algorithms (such as Q-learning, DDPG, and PPO) to construct a state space that includes forecasts for photovoltaic power generation, air conditioning load, electricity price information, and energy storage status, as well as an action space for energy storage charge and discharge power decisions. By maximizing the objective function of long-term comprehensive benefits, it learns and outputs the optimal energy storage scheduling strategy, thus balancing economic efficiency and system stability, effectively improving the overall system operational efficiency.

[0185] Based on the results of the prediction model output and user preference settings, the intelligent control module integrates multi-source data and edge computing capabilities to achieve intelligent decision-making on energy flow paths, adaptive adjustment of air-conditioning operating status, and continuous optimization of the overall system operating efficiency.

[0186] The intelligent control module includes an energy flow adaptive decision-making unit and an air conditioning operation optimization unit. The energy flow adaptive decision-making unit dynamically analyzes multi-dimensional time-series data, including photovoltaic power generation, air conditioning load, electricity price signals, and energy storage status, to build a state-aware and rule-driven energy flow allocation rules engine. This engine adjusts energy path priorities according to the predefined rules in the core rule library in Table 1, implementing a tiered energy supply strategy that prioritizes direct photovoltaic supply, balances energy storage, and provides a backup grid.

[0187] The startup energy storage protection strategy is introduced to set the energy storage system start-stop logic and minimum working current protection to prevent frequent start-stop and system life degradation.

[0188] The energy storage system start-up and shutdown logic includes:

[0189] If the energy storage needs to switch from charging to discharging and the first start-stop switching condition is met, the energy storage is discharged; the first start-stop switching condition is to meet the following formula within the set anti-shake delay:

[0190] P c <P c_on_threshold +P hys

[0191] Among them, P c is the actual charging power of energy storage; P c_on_threshold The required energy storage charging power; P hys is the hysteresis interval; preferably, the required energy storage charging power is -0.5kW; the preferred value range of the hysteresis interval is ±0.2kW, charging is negative and discharging is positive; the anti-shake delay is set to less than or equal to 30 seconds.

[0192] If the energy storage needs to switch from discharging to charging and the second start-stop switching condition is met, the energy storage is charged.

[0193] The second switching condition after the start-up and stagnation is to satisfy the following formula within the set anti-shake delay:

[0194] P l ≥P l_on_threshold +P hys

[0195] Among them, P l is the actual discharge power of energy storage; P l_on_threshold is the required energy storage discharge power; P hys is the hysteresis interval; preferably, the required energy storage discharge power is 0.5kW; the preferred value range of the hysteresis interval is ±0.2kW, charging is negative and discharging is positive; the anti-shake delay is set to less than or equal to 30 seconds.

[0196] The first start-stop switching condition and the second start-stop switching condition are both set based on the actual battery power and the required battery power.

[0197] The minimum operating current protection is to only perform charging and discharging when the energy storage charging and discharging current is greater than or equal to the set minimum operating current threshold, otherwise it remains idle. The function of the minimum operating current protection is to eliminate small current operations and reduce battery loss. Therefore, those skilled in the art should be able to set the minimum operating current threshold according to the actual energy storage capacity; for example, when the energy storage capacity is 20Ah, the minimum operating current threshold is set to 2A; when the energy storage capacity is 30Ah, the minimum operating current threshold is set to 3A.

[0198] Based on the scheduling execution agent mechanism of the edge gateway, the power supply strategy is adjusted periodically according to the set first time interval to achieve real-time dynamic adjustment of energy flow. Preferably, the set first time interval is 1 minute;

[0199] The air conditioning operation optimization unit, based on a fuzzy PID controller, adaptively adjusts the operating status of the air conditioning compressor, aiming to precisely control indoor temperature and humidity while dynamically optimizing the air conditioning system's energy consumption. This unit integrates multi-source environmental and operational data with predictive model results to achieve intelligent and efficient air conditioning control. Specifically, the fuzzy PID controller uses "temperature error," "error change rate," and "humidity feedback" as fuzzy input variables, and utilizes fuzzy logic reasoning and a rule base to generate control outputs. The control signal is ultimately converted into a PWM duty cycle, which is used to adjust the compressor speed for fine-tuning cooling or heating intensity. The fuzzy set partitioning and rule design within the controller are optimized based on the actual system operating conditions and user experience requirements, ensuring a smooth and responsive control process. The temperature error represents the instantaneous deviation between the user-set temperature value and the actual measured indoor temperature; the error change rate represents the rate of change of the temperature error over time; and the humidity feedback represents the deviation between the user-set humidity value and the actual measured humidity.

[0200] The preset fuzzy rule base includes:

[0201] When the temperature error is greater than a set first temperature error threshold, the error change rate is greater than a set first change rate threshold, and the indoor humidity is greater than the user-set humidity, the air conditioner compressor speed is reduced to a set first speed; the temperature error is the difference between the user-set temperature and the indoor temperature; the error change rate is the rate of change of the temperature error;

[0202] When the temperature error is less than the set second temperature error threshold, the error change rate is 0, and the indoor humidity is equal to the user-set humidity, the current air-conditioning compressor speed is maintained;

[0203] When the temperature error is negative and greater than the set third temperature error threshold, and the error change rate is greater than the set second change rate threshold, and the indoor humidity is lower than the user-set humidity, the air conditioner compressor speed is increased to the set second speed;

[0204] When the temperature error is negative and less than a set fourth temperature error threshold, and the error change rate is less than a set second change rate threshold, increasing the air conditioner compressor speed to a set third speed;

[0205] The first temperature error threshold is greater than the second temperature error threshold; the first temperature error threshold, the second temperature error threshold, and the first rate of change threshold are all positive numbers; those skilled in the art can set the first temperature error threshold, the second temperature error threshold, and the first rate of change threshold based on actual conditions, and will not be described in detail here. The set first speed is based on the current air conditioner compressor speed. Those skilled in the art can set the first speed to be less than the current air conditioner compressor speed based on actual conditions, and will not be described in detail here.

[0206] The third temperature error threshold is greater than the fourth temperature error threshold; the third temperature error threshold, the fourth temperature error threshold, and the second rate of change threshold are all negative; the ratio of the second speed to the current air-conditioning compressor speed is less than the ratio of the third speed to the current air-conditioning compressor speed. Those skilled in the art can set the third temperature error threshold, the fourth temperature error threshold, and the second rate of change threshold based on actual conditions, and will not be described in detail here. The set second and third speeds are based on the current air-conditioning compressor speed. Those skilled in the art can set the second and third speeds based on actual conditions, and will not be described in detail here.

[0207] The air conditioning operation optimization unit supports multiple operation mode switching. Energy-saving mode allows the error range of indoor temperature control and the user-set preferred temperature to be ±3°C, reducing the frequent start and stop times of the compressor, thereby reducing power consumption and equipment wear, and extending system life. Comfort mode reduces the error range of temperature control and the user-set preferred temperature to ±1°C to enhance user comfort experience, and is suitable for scenarios with high requirements for indoor environment. Mode switching can be automatically executed based on factors such as the current operating mode set by the user, the comfort mode time period set by the user, the power consumption budget, and the air conditioning load, to achieve a dynamic balance between power consumption and comfort. Specifically, first, weights are assigned to the current operating mode, power consumption budget, air conditioning load, and comfort mode time period set by the user, the mode switching coefficient is calculated, and the operating mode is determined based on the mode switching coefficient; the mode switching coefficient C O As shown in the following formula:

[0208]

[0209] Among them, a is the weight of the operating mode; R uThe current operating mode set by the user. When the current operating mode is energy saving mode, R u When the current operating mode is comfort mode, R u is 2; b is the power consumption budget weight; E b The energy consumption budget for the second time period is set (in kW.h); E n is the power consumption of the air conditioner in the first set period (in kW.h); preferably, the second period is set to 1 day; c is the air conditioner load weight; A l is the air conditioning load (in kW); A e is the rated load of the air conditioner (in kW); d is the weight of the comfort mode time period; T s The length of comfort mode time set by the user, in h; T e The first time length standard is set, preferably, the first time length standard is 24 hours.

[0210] The operating mode weight, power consumption budget weight, air conditioning load weight, and comfort mode time period weight must meet the following constraints:

[0211]

[0212] Preferably, the weight of the operating mode is 0.4; the weight of the power consumption budget is 0.3; the weight of the air-conditioning load is 0.2; and the weight of the comfort mode time period is 0.1.

[0213] When the mode switching coefficient is greater than the set switching coefficient threshold, the comfort mode is adopted, otherwise, the energy-saving mode is adopted; those skilled in the art should be able to set the switching coefficient threshold according to actual conditions, which will not be repeated here.

[0214] In addition, the air-conditioning operation optimization unit combines the air-conditioning load prediction results output by the prediction and optimization modeling module to adjust the control strategy in advance. For example, before the load peak is expected to occur, the compressor operation intensity is increased in advance to avoid frequent start and stop under high load, thereby improving the system response flexibility and overall energy efficiency; when it is predicted that there will be strong sunlight in the afternoon (according to the weather forecast), the system will combine the photovoltaic output and the air-conditioning load forecast to moderately pre-cool the indoor environment in the morning stage, so that the air-conditioning system can complete the pre-cooling before the heat load surges, significantly reducing the pressure of electricity consumption under peak load. Those skilled in the art can determine the proportion of increasing the compressor operation intensity according to actual conditions, or they can determine the minimum temperature for pre-cooling the indoor environment according to the actual operation mode, the temperature and humidity set by the user, which will not be elaborated here.

[0215] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0216] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0217] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0218] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0219] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A photovoltaic energy storage air conditioner intelligent control method, characterized in that: include: Collecting photovoltaic energy storage air conditioning related data; the photovoltaic energy storage air conditioning related data includes photovoltaic related data, energy storage related data, and air conditioning operation related data; Input photovoltaic related data into the photovoltaic power generation prediction model to obtain the photovoltaic power generation prediction result; input air conditioning operation related data into the air conditioning load prediction model to obtain the air conditioning load prediction result; The photovoltaic power generation forecast results, air conditioning load forecast results, energy storage related data, and the historical photovoltaic power generation data and air conditioning load data within the set first time period are input into the energy storage optimization scheduling model to obtain energy storage charging and discharging decisions; Based on energy storage-related data, peak electricity price periods, and the output results of the photovoltaic power generation forecast model and the air conditioning load forecast model, a power supply strategy core rule base is constructed; based on the power supply strategy core rule base, the output results of the energy storage optimization scheduling model are optimized.

2. The photovoltaic energy storage air conditioner intelligent control method according to claim 1, characterized in that: The method further includes: constructing a compressor state adjustment fuzzy rule base based on set user preferences and temperature and humidity values ​​in air conditioning operation-related data; and implementing adaptive adjustment of the air conditioning compressor operation state through a fuzzy controller based on the compressor state adjustment fuzzy rule base.

3. The photovoltaic energy storage air conditioner intelligent control method according to claim 1 is characterized in that: Photovoltaic related data include photovoltaic DC side output voltage, photovoltaic DC side output current, photovoltaic module temperature, outdoor irradiance, outdoor temperature, outdoor humidity and wind speed; Air conditioning operation related data includes compressor frequency, condenser temperature, evaporator temperature, indoor temperature, indoor humidity and power consumption; Energy storage related data includes the voltage of each single cell, the temperature of each single cell, the charge and discharge current of the energy storage, the energy storage status, the charge status, the health status and the charge and discharge rate; the energy storage status includes energy storage charging, discharging and neither charging nor discharging.

4. The photovoltaic energy storage air conditioner intelligent control method according to claim 1, characterized in that: The power supply strategy core rule base includes the first core rule, the second core rule and the third core rule; Among them, the first core rule has a higher priority than the second core rule; the second core rule has a higher priority than the third core rule; The first core rule is that when the predicted photovoltaic power generation is greater than or equal to the predicted air conditioning load, photovoltaic power is used to directly power the air conditioning, and the remaining power is used to charge the energy storage. The second core rule is that when the sum of the predicted photovoltaic power generation and the energy stored in the energy storage is greater than or equal to the predicted air conditioning load, photovoltaic power is used directly to power the air conditioning, and energy storage discharge is used to supplement the power supply difference; The third core rule is that when electricity prices are at peak times and the energy storage state of charge is greater than the set charge threshold, energy storage is forced to power the air conditioner.

5. The photovoltaic energy storage air conditioner intelligent control method according to claim 1, characterized in that: When optimizing the output results of the energy storage optimization scheduling model, the preset energy storage protection strategy is executed; The energy storage protection strategy includes energy storage system start-up and stall logic and minimum operating current protection.

6. The photovoltaic energy storage air conditioner intelligent control method according to claim 5, characterized in that: The energy storage system start-up and shutdown logic includes: If the energy storage needs to switch from charging to discharging and the first start-stop switching condition is met, the energy storage is discharged; If the energy storage needs to switch from discharging to charging and the second start-stop switching condition is met, the energy storage is charged.

7. The photovoltaic energy storage air conditioner intelligent control method according to claim 6, characterized in that: The first switching condition after the start-up stagnation is to satisfy the following formula within the set anti-shake delay: P c <P c_on_threshold +P hys Among them, P c is the actual charging power of energy storage; P c_on_threshold The required energy storage charging power; P hys is the hysteresis interval.

8. The photovoltaic energy storage air conditioner intelligent control method according to claim 6 or 7, characterized in that: The second switching condition after the start-up and stagnation is to satisfy the following formula within the set anti-shake delay: P l ≥P l_on_threshold +P hys Among them, P l is the actual discharge power of energy storage; P l_on_threshold is the required energy storage discharge power.

9. The photovoltaic energy storage air conditioner intelligent control method according to claim 5, characterized in that: The minimum operating current protection is to perform charging and discharging only when the energy storage charging and discharging current is greater than or equal to the set minimum operating current threshold, otherwise it remains idle.

10. The photovoltaic energy storage air conditioner intelligent control method according to claim 2, characterized in that: The compressor state adjustment fuzzy rule base includes: When the temperature error is greater than a set first temperature error threshold, the error change rate is greater than a set first change rate threshold, and the indoor humidity is greater than the user-set humidity, the air conditioner compressor speed is reduced to a set first speed; the temperature error is the difference between the user-set temperature and the indoor temperature; the error change rate is the rate of change of the temperature error; When the temperature error is negative and greater than the set second temperature error threshold, and the error change rate is less than the set second change rate threshold, and the indoor humidity is less than the user-set humidity, the air conditioner compressor speed is increased to the set second speed; When the temperature error is negative and less than a set second temperature error threshold, increasing the speed of the air-conditioning compressor to a set third speed; In other cases, the current air conditioner compressor speed is maintained; a first temperature error threshold; the first temperature error threshold and the first change rate threshold are both positive numbers; The second temperature error threshold and the second change rate threshold are both negative numbers; the ratio of the second speed to the current air-conditioning compressor speed is less than the ratio of the third speed to the current air-conditioning compressor speed.

11. A photovoltaic energy storage air conditioning intelligent control system using the photovoltaic energy storage air conditioning intelligent control method according to any one of claims 1 to 10, characterized in that: It includes multi-source data acquisition module, prediction and optimization modeling module and intelligent control module: The multi-source data acquisition module collects photovoltaic energy storage air conditioning related data; the photovoltaic energy storage air conditioning related data includes photovoltaic related data, energy storage related data, and air conditioning operation related data; The prediction and optimization modeling module inputs photovoltaic-related data into the photovoltaic power generation prediction model to obtain photovoltaic power generation prediction results; inputs air conditioning operation-related data into the air conditioning load prediction model to obtain air conditioning load prediction results; and inputs photovoltaic power generation prediction results, air conditioning load prediction results, energy storage-related data, and photovoltaic power generation historical data and air conditioning load historical data within a set first time period into the energy storage optimization scheduling model to obtain energy storage charging and discharging decisions; The intelligent control module constructs a power supply strategy core rule base based on energy storage-related data, peak electricity price periods, and the output results of the photovoltaic power generation prediction model and the air conditioning load prediction model; optimizes the output results of the energy storage optimization scheduling model based on the power supply strategy core rule base; and constructs a compressor state adjustment fuzzy rule base based on the set user preferences and the temperature and humidity values ​​in the air conditioning operation-related data; Based on the compressor state adjustment fuzzy rule base, the adaptive adjustment of the operating state of the air-conditioning compressor is achieved through a fuzzy controller.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the photovoltaic energy storage air conditioner intelligent control method according to any one of claims 1 to 10 are implemented.

Images